A High Voltage Direct Current (HVDC) power transmission system is a viable alternative to alternating current (AC) power transmission systems, for example for long-distance distribution. Up to date, HVDC power transmission has been point-to-point two terminal power transmission with a few exceptions. A multi-terminal HVDC power transmission is more complex than the ordinary point-to-point power transmission. In particular, the control system is more elaborate and telecommunication requirements between stations become larger. A major reason for the more elaborate control system is difficulties to control the power flow within a large HVDC system, especially at disturbances.
In case of disturbances within the HVDC power transmission network, for example outages of converters or lines, control actions have to be taken in order to ensure stability of the network and the power distribution. The aim of any such control actions is to handle the disturbances and provide a distribution of load that is acceptable and that enables uninterrupted power to be delivered to end users.
Droop control is a well known method for handling disturbances. Such droop control is described in, for example, “Control of Multiterminal HVDC Transmission for Offshore Wind Energy”, T. Haileselassie et al. The document is mainly aimed at means for avoiding the need for communication when controlling the power distribution at a disturbance. Although functioning for load distribution at large disturbances such as loss of a converter, a drawback of the described droop control method is difficulties handling minor disturbances. In particular, even a rather small error in measurement will give great impact on the whole system.
Further, the method described could operate acceptably for a star network configuration, but would encounter difficulties for other types of network configurations.
In view of the above, there is a need for a solution that ensures proper control of multi-terminal DC power transmission systems during any type of disturbance as well as during steady state operation. Further, it would be desirable to provide such solution usable for any type of network configuration.